Google DeepMindproprietary
Released: 2026-10-28Artificial AnalysisGemini 4 Carbon
Gemini 4 Carbon is a frontier Gemini 4 model from Google for coding, agentic and complex reasoning work. It takes text, image, audio, video and file input, supports long decode...
Intelligence Index
60.0/ 100
Artificial Analysis Intelligence Index (independent composite)Throughput Speed
61tok/sec
Output streaming throughputTime to First Answer
72s
Time to first answer token, including any thinkingCost to Run the Index
5.48USD
Artificial Analysis cost of the full Index runMeasured results · Artificial Analysis
Independent evaluations of Gemini 4 Carbon (High configuration). Values are reported as published; nothing here is estimated or derived. Snapshot 2026-11-15.
- Intelligence Index
- 60.0
- Cost to run the Index
- $5.48
- Time per task
- 33.8 min
- Omniscience index
- 53.5
- Reasoning model
- Yes
- Cached-input price
- $0.1 / 1M
Reasoning & knowledge
- AA-Omniscience Accuracy58.3%
- CritPt v2 (Hard)69.0%
- HLE-Diamond69.1%
Coding & terminal agents
Agentic work
- AA-Briefcase v1.1 (Elo)1611 Elo
- AutomationBench-AA77.8%
- GDPval-AA (normalized Elo)1733 Elo
- Multimodal Knowledge Agent (MKA)54.3%
Long context & documents
Multimodal & science
- VIALS-AA64.0%
Health & safety
Reasoning-effort variants evaluated
| Variant | Intelligence Index | Cost to run Index | Output speed (tok/s) |
|---|---|---|---|
| High | 60.0 | $5.48 | 60.6 |
| Medium | 58.3 | $4.68 | 59.7 |
| Low | 54.2 | $5.22 | 51.2 |
Headline Evaluation Breakdown
Six independently measured evaluations (Artificial Analysis)Reasoning · CritPt v2 (Hard)69.0%
Coding & agents · Terminal-Bench 4.065.1%
Long context · AA-LCR79.3%
Sourced Benchmark Evaluations (13)
Verified performance across authoritative benchmarks with provenance tracking.| Benchmark | Status | Score | Trust | Date | Source Type | Provenance |
|---|---|---|---|---|---|---|
| Terminal-Bench 4.0 | Active | 65.2% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench-Science 0.1 | Active | 64.8% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-LCR (Long Context Reasoning) | Active | 79.3% | 93 | 2026-11-15 | independent | Source ↗ |
| CritPt v2 (Hard) | Active | 69% | 93 | 2026-11-15 | independent | Source ↗ |
| HLE-Diamond | Active | 69.1% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-Omniscience (Accuracy) | Active | 58.3% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-Omniscience (Non-Hallucination) | Active | 88.6% | 93 | 2026-11-15 | independent | Source ↗ |
| FrontierCode-AA 2.0 | Active | 40.6% | 93 | 2026-11-15 | independent | Source ↗ |
| GDP.pdf | Active | 28.4% | 93 | 2026-11-15 | independent | Source ↗ |
| MLCR-AA (Medical Long Context Reasoning) | Active | 19.4% | 93 | 2026-11-15 | independent | Source ↗ |
| Multimodal Knowledge Agent (MKA) | Active | 54.3% | 93 | 2026-11-15 | independent | Source ↗ |
| AutomationBench-AA | Active | 77.8% | 93 | 2026-11-15 | independent | Source ↗ |
| VIALS-AA | Active | 64% | 93 | 2026-11-15 | independent | Source ↗ |
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="google-gemini-4-carbon",
messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
temperature=0.2
)
print(response.choices[0].message.content)Evaluation & Sourcing Notes
Scores listed for Gemini 4 Carbon represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.
All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.